Local LLM model page

Qwen 2.5 Coder (32B)

Best open-source code model at 32B. Outstanding code generation, completion and refactoring. 7.5M downloads.

Find the best model for my hardware Browse all 183 LLMs

Parameters

32B

Minimum RAM

24 GB

Model size

19 GB

Quantization

Q5_K_M

Can Qwen 2.5 Coder (32B) run locally?

Qwen 2.5 Coder (32B) is best suited for power-user machines with 32 GB RAM. LocalClaw recommends Q5_K_M as the default quantization, with at least 24 GB RAM.

Search term for LM Studio or compatible runtimes: qwen2.5-coder-32b-instruct

Hugging Face repository: lmstudio-community/Qwen2.5-Coder-32B-Instruct-GGUF

codepowerquality

Strengths

Best open-source code model
7.5M downloads
128K context
Rivals GPT-4 for coding
Apache 2.0

Limitations

Needs 24GB+ RAM
Coding specialist — not for general tasks

Best use cases

Professional software development
Code review
Architecture design
Legacy code migration

Benchmarks

Speed: 4/10

Quality: 8/10

Coding: 10/10

Reasoning: 8/10

Technical details

Developer: Alibaba Cloud (Qwen Team)

License: Apache 2.0

Context window: 131,072 tokens

Architecture: Transformer fine-tuned for code

Released: 2024-11